Commit message (Collapse) | Author | Age | Lines | |
---|---|---|---|---|
* | Merging mainline. | Wilmer van der Gaast | 2010-06-06 | -129/+164 |
|\ | ||||
| * | Disable old-style ICQ authentication. It looks like AOL or whoever is slowly | Wilmer van der Gaast | 2010-06-02 | -1/+3 |
| | | | | | | | | | | | | taking down support for it. Just to be sure, it can be re-enabled with a setting, I'll remove that after the next release. | |||
| * | Merge from Geert: Added ability to follow/unfollow Twitter people. (Unfollow | Wilmer van der Gaast | 2010-05-30 | -5/+16 |
| |\ | | | | | | | | | | | | | won't work with mode=one.) | |||
| | * | Added functionality to add and remove friendships. | Geert Mulders | 2010-05-29 | -5/+21 |
| | | | ||||
| * | | Restructure Twitter error parser a bit, it fed a NULL pointer to the XML | Wilmer van der Gaast | 2010-05-23 | -17/+16 |
| | | | | | | | | | | | | | | | parser sometimes (which fails safely but is a bad idea anyway). | |||
| * | | Don't whine on "incomplete" incoming IQ-set packets either, those also happen. | Wilmer van der Gaast | 2010-05-23 | -1/+0 |
| | | | ||||
| * | | Improved error reporting (get textual HTTP error message and error message | Wilmer van der Gaast | 2010-05-23 | -12/+36 |
| | | | | | | | | | | | | | | | from Twitter API response if possible). | |||
| * | | Added base_url settting to Twitter module so other services using the | Wilmer van der Gaast | 2010-05-23 | -108/+99 |
| | | | | | | | | | | | | | | | Twitter API can be used. Only with Basic authentication though. | |||
| * | | Use HTTPS for the browser part of Twitter OAuth login. | Wilmer van der Gaast | 2010-05-22 | -1/+1 |
| | | | ||||
| * | | Setting the HTML flag on a connection has a nasty side effect of escaping | Wilmer van der Gaast | 2010-05-19 | -1/+3 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | a lot of "special" characters, and these HTML entities are not counted as one character. :-( So just strip HTML of incoming stuff and don't do anything with what goes out. It's not required. The story may actually be more complicated this, let's find out. | |||
| * | | Oi, don't leak that precious memory! | Wilmer van der Gaast | 2010-05-16 | -1/+2 |
| | | | ||||
| * | | Handle AIM_CAPS_ICQRTF messages (sometimes sent by certain mobile clients). | Wilmer van der Gaast | 2010-05-16 | -2/+6 |
| | | | | | | | | | | | | | | | Not sure if this needs any stripping but I guess we'll find out. | |||
| * | | Passport 3.0 authentication turns out to be slightly different from what | Wilmer van der Gaast | 2010-05-15 | -1/+3 |
| | | | | | | | | | | | | | | | | | | the msnpiki MSNP13 doc says and stuff is breaking now. This should fix the main problem now. | |||
* | | | Merging libpurple branch into killerbee. It's fairly usable already, and | Wilmer van der Gaast | 2010-06-06 | -287/+3666 |
|\ \ \ | | | | | | | | | | | | | | | | | | | | | Debian packaging is now properly separated. This also picks up a load of stuff from mainline it seems. | |||
| * | | | Allow one to run the configure script from a different directory and put all | Wilmer van der Gaast | 2010-05-25 | -7/+29 |
| | | | | | | | | | | | | | | | | | | | | | | | | build files in there. I need this to properly make Debian package variants (i.e. libpurple and native). | |||
| * | | | Merging killerbee. (The non-libpurple stuff currently doesn't compile.) | Wilmer van der Gaast | 2010-05-25 | -1/+1 |
| |\ \ \ | ||||
| * | | | | GAIM_INPUT_* were renamed, at last. | Wilmer van der Gaast | 2010-05-25 | -13/+13 |
| | | | | | ||||
| * | | | | Read display names. Setting them is going to be an awesome hack. | Wilmer van der Gaast | 2010-05-24 | -0/+22 |
| | | | | | ||||
| * | | | | Support for named groupchats, although not very solid. | Wilmer van der Gaast | 2010-05-22 | -8/+44 |
| | | | | | ||||
| * | | | | Enable changing and viewing of block/allow lists. | Wilmer van der Gaast | 2010-05-22 | -0/+75 |
| | | | | | ||||
| * | | | | Return ui_info so jabber:iq:version responses will not say just libpurple. | Wilmer van der Gaast | 2010-05-22 | -0/+15 |
| | | | | | ||||
| * | | | | Added support for the info command. | Wilmer van der Gaast | 2010-05-22 | -0/+62 |
| | | | | | ||||
| * | | | | Sigh. Enable debugging only if the BITLBEE_DEBUG variable is set. | Wilmer van der Gaast | 2010-05-22 | -1/+3 |
| | | | | | ||||
| * | | | | Fixed sending with proper filenames by creating a temporary directory with | Wilmer van der Gaast | 2010-05-21 | -23/+43 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the file in it; protocol modules are mostly hardcoded to use the filename from the filesystem with no way to override this. Also improved robustness a little bit. | |||
| * | | | | Similar hacky code to send files. This indirect sending stuff sucks badly | Wilmer van der Gaast | 2010-05-19 | -20/+93 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | for numerous reasons. Maybe libpurple 2.7.0 is less crappy and will eventually allow (working) direct ft's again. This somewhat works, but filename info is lost with some protocols. | |||
| * | | | | Suppress auto-reconnect when required (auth errors and concurrent logins | Wilmer van der Gaast | 2010-05-18 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | | | | | | | probably, not sure what sets the wants_to_die flag). | |||
| * | | | | Little cleanup. Less compiler warnings, and removing tempfile at the | Wilmer van der Gaast | 2010-05-18 | -80/+85 |
| | | | | | | | | | | | | | | | | | | | | | | | | | beginning of the download already to make sure it doesn't stick around. | |||
| * | | | | purple_conv_chat_invite_user() is libpurple >= 2.6.0, so use | Wilmer van der Gaast | 2010-05-18 | -1/+4 |
| | | | | | | | | | | | | | | | | | | | | | | | | | serv_chat_invite() instead. | |||
| * | | | | This receives files but is very fragile if anything unusual happens (like a | Wilmer van der Gaast | 2010-05-18 | -15/+102 |
| | | | | | | | | | | | | | | | | | | | | | | | | | cancellation/timeout/etc). | |||
| * | | | | Move direct ft stuff to an unused file: This gets too hairy and too fragile. | Wilmer van der Gaast | 2010-05-17 | -101/+275 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I don't have time to work out all the details, I doubt if this is supposed to work reliably yet at all. Let's go for the simple via-disk approach for now. | |||
| * | | | | Split off the file transfer stuff into a separate file. What a mess. | Wilmer van der Gaast | 2010-05-17 | -198/+237 |
| | | | | | ||||
| * | | | | Mainline merge. (Probably mostly irrelevant for this branch, oh well.) | Wilmer van der Gaast | 2010-05-16 | -209/+271 |
| |\ \ \ \ | | | |/ / | | |/| | | ||||
| | * | | | Add a warning when the user included a domain part in his/her Yahoo! username. | Wilmer van der Gaast | 2010-05-14 | -0/+5 |
| | | | | | ||||
| | * | | | Stop those "Many switchboard failures" messages. No point in showing them | Wilmer van der Gaast | 2010-05-11 | -27/+22 |
| | | | | | | | | | | | | | | | | | | | | | | | | | unless the lost switchboard actually had something queued. | |||
| | * | | | Fixed a long-standing bug where non-anonymous Jabber chatrooms could cause | Wilmer van der Gaast | 2010-05-11 | -1/+3 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | full JIDs to be added to the local contact list, but not removed when leaving the room, eventually triggering errors like the one in #499. People kept sending me these dumps and I finally took time to make what turned out to be this one-line fix. This error should never appear again unless people are doing really crazy stuff. | |||
| | * | | | It looks like AIM mem requests (implemented as damn_you() and | Wilmer van der Gaast | 2010-05-09 | -136/+0 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | straight_to_hell() back in the days - and still more or less the same in libpurple it seems) are no longer necessary; the script no longer exists at the location used in the source and nobody complained. Nice try, AOL. And hats of to the Gaim developer who thought of this clever way around it. | |||
| | * | | | Room names on OSCAR can't start with digits, this broke the "chat with" | Wilmer van der Gaast | 2010-05-09 | -1/+2 |
| | | | | | | | | | | | | | | | | | | | | | | | | | command on ICQ so far. Just prepend "icq_" and it'll work. | |||
| | * | | | Yahoo! seems to echo conference room invitations now, huh? No idea what this | Wilmer van der Gaast | 2010-05-09 | -0/+4 |
| | | |/ | | |/| | | | | | | | | | | | | | is but let's ignore them. | |||
| | * | | Check Tweet length on the BitlBee side already. | Wilmer van der Gaast | 2010-05-03 | -2/+16 |
| | | | | ||||
| | * | | OAuth sanity fix: Twitter-specific stuff should *not* be in lib/oauth.c. | Wilmer van der Gaast | 2010-05-02 | -33/+40 |
| | |\ \ | | | | | | | | | | | | | | | | | | | | | Somewhat intrusive, should've done this right immediately. :-/ | |||
| | | * | | Save the credentials again. | Wilmer van der Gaast | 2010-05-01 | -4/+4 |
| | | | | | ||||
| | | * | | Cleaned up OAuth stuff: consumer key/secret should *not* be in lib/oauth.c. | Wilmer van der Gaast | 2010-05-01 | -35/+42 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Keep it in the Twitter module, and use the oauth_info struct through the whole session to keep all this together. | |||
| | * | | | Applied show-offline patch from Florian E.J. Fruth, adapted for a few | Wilmer van der Gaast | 2010-05-02 | -16/+74 |
| | | | | | | | | | | | | | | | | | | | | | | | | | changes that happened since 1.2.4. | |||
| | * | | | Oops. rt_mon start at 0=Jan, causing off-by-a-month timestamps shown to the | Wilmer van der Gaast | 2010-05-01 | -1/+1 |
| | |/ / | | | | | | | | | | | | | | | | | user. Temporary work-around: set timezone to +720 or so. :-P | |||
| | * | | Set HTML-flag on Twitter accounts so </> and all get converted back. | Wilmer van der Gaast | 2010-04-30 | -0/+1 |
| | | | | | | | | | | | | | | | | | | | | (Probably also fixes potential problems with outgoing messages.) | |||
| | * | | Valgrind cleanup. | Wilmer van der Gaast | 2010-04-27 | -9/+21 |
| | | | | ||||
| | * | | Save the OAuth token in the acct structs so it doesn't have to be rerequested | Wilmer van der Gaast | 2010-04-26 | -0/+5 |
| | | | | | | | | | | | | | | | | | | | | every time. | |||
| | * | | OAuth, it lives! | Wilmer van der Gaast | 2010-04-26 | -3/+37 |
| | | | | ||||
| | * | | Twitter module now generates authorize URLs. | Wilmer van der Gaast | 2010-04-26 | -14/+54 |
| | | | | ||||
| | * | | Successfully posted a tweet! | Wilmer van der Gaast | 2010-04-26 | -9/+29 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Twitter's tricky. It returns vars (user_id, screen_name) in the access token that, the way I read the spec, should be included in all subsequent queries. However, stuff only started to work when I dropped those vars. This code's definitely not pretty ATM. Need to clean up now that it actually works. |